Experimental IRC log swig-2009-06-05

Available formats: content-negotiated html turtle (see SIOC for the vocabulary)

Back to channel and daily index: content-negotiated html turtle

These logs are provided as an experiment in indexing discussions using IRCHub.py, Irc2RDF.hs, and SIOC.

07:25:46<mhausenblas>morning Web Of Data
07:38:23<EtnaRosso>salve a tutti
07:58:18<phrearch>morning
08:09:02<danbri>mornin'
08:13:43<lmaurits>Morning!
11:36:32<danbri>http://walking-papers.org/
11:36:45<danbri>A:|Paper-based map annotation!
11:36:50<danbri>A:Very nice
11:36:57<danbri>A:"Print maps, draw on them, scan them back in and help OpenStreetMap improve its coverage of local points of interests and street detail."
13:50:05<MacTed>phenny, ask pipian to please add http://telegraphis.net/data/ dumps and other info to http://esw.w3.org/topic/DataSetRDFDumps , http://esw.w3.org/topic/TaskForces/CommunityProjects/LinkingOpenData/DataSets/LinkStatistics , http://esw.w3.org/topic/TaskForces/CommunityProjects/LinkingOpenData/DataSets/Statistics , http://esw.w3.org/topic/TaskForces/CommunityProjects/LinkingOpenData/DataSets , etc.
13:50:05<phenny>MacTed: I'll pass that on when pipian is around.
16:16:29<tobyink>89 hours until awesomeness arrives.
16:20:18<_psychic_>tobyink: 89 hours?
16:20:34<tobyink>Indeed. Slightly less in fact.
16:20:41<_psychic_>How so?
16:21:20<tobyink>Well, awesomeness arrives at 9:00Z on Tuesday, and that's less than 89 hours away.
16:24:24<_psychic_>Okay. :/
16:27:03<tobyink>Well, if I said any more it would spoil the surprise.
16:30:31<_psychic_>I suppose that's what I was hoping for, in a way. :)
16:31:58<csarven>tobyink Is awesomeness for all mankind or just you? =)
16:33:00<tobyink>For the subset of mankind with Internet access.
17:07:44<mattl>tobyink: are you bringing awesomeness?
17:10:17<tobyink>Yes, on Tuesday.
17:11:57<melvster>anyone know how well sqlight works as a triplestore?
17:12:19<PovAddict>sqlite?
17:13:33<melvster>sorry yes
17:14:29<melvster>i was thinking of putting the swse deri dump in there, but now i cant find it
17:18:46<melvster>hmm not linked in anymore, but still here: http://sw.deri.org/2009/01/visinav/faq.html
17:33:46<SifuMoraga>I just posted an unprocessed video from last night's Semantic Web Meetup on Semantic Advertising.
17:33:49<SifuMoraga>It's rough.
17:33:53<SifuMoraga>It's ugly.
17:33:57<SifuMoraga>It's lowres.
17:34:01<SifuMoraga>But it's up: http://vimeo.com/5019681
17:34:08<akozak>semantic advertising?
17:34:30<SifuMoraga>Yes, that was my thought, too. But it was more interesting than expected.
17:39:11<melvster>always thought semantic advertising was a pretty natural idea, look at items in your foaf:interests, then present the best value products in that field
17:48:52<SifuMoraga>They were talking about how to do better advertising with semantics
17:49:30<akozak>man this video has a long intro
17:54:47<rreck>if i would like to make an assertion about a paragraph/component of a document does anyone know an elegant way to refer (URI) to it?
17:57:36<akozak>can't you use html anchors?
17:58:14<rreck>yes, if i wanted to edit the source. im talking about my facebook page. imagine i wanted to say something about "my contact information" only
17:58:52<rreck>i dont think there is an "easy" answer but i thought id ask before making something up
18:03:22<rreck>maybe i could do something with <div id=\"info_section_contact\" class=\"info_section\"> ?
18:03:58<rreck>id != #name
18:18:50<danbri>anyone with python powers got a minute?
18:18:59<danbri>am trying to do something that i'd find easy in perl or ruby ...
18:19:02<danbri> def codelink(self, s):
18:19:02<danbri> reg1 = re.compile(r"""<code>foaf:(\w+)<\/code>""")
18:19:02<danbri> s_linked = reg1.replace(r"""<code><a href=\"#term_#\1\">foaf:#\1</a></code>""", s)
18:19:03<danbri> return(s_linked)
18:19:47<danbri>it goes from <code>foaf:xyz</code> to <code><a href="#term_xyz">foaf:xyz</a></code>
18:19:55<danbri>well at the moment it doesn't do anything :)
18:57:19<melvster>rreck: dont know a definitive answer but does xpath help?
18:58:46<melvster>there's a firefox addon called autopager with mades asserstions about tags in a page using xpath and stores that in XML, in particular, it finds the next button on almost any page
18:59:43<PovAddict>ugh
18:59:46<melvster>it also uses xpath to delimit the 'content' in the page
18:59:50<PovAddict><link rel="next">!
19:00:08<melvster>not all pages have that, and you also need to know which part of the next page to pull
19:00:10<PovAddict>few websites use that <link>, but also, I wish browsers *used* it when it's present
19:00:20<melvster>opera uses it
19:00:25<melvster>just press space
19:00:32<PovAddict>Konqueror has an "up" button
19:00:44<PovAddict>that removes the last slash-separated part of the URL
19:00:49<PovAddict>instead of using the <link> taking "up"
19:01:00<melvster>PovAddict: you should try autopager, it's awesome
19:01:23<melvster>i dream of a global rdfa version
19:01:34<PovAddict>probably because Konqueror's it's-a-webbrowser-it's-a-file-browser-it's-BOTH is more even extreme than Windows
19:01:57<melvster>i rarely use anything other than opera as its ultra fast
19:02:04<rreck>melvster: ty
19:02:25<melvster>rreck, sorry i dont know the answer, but it's a very interesting question
19:02:53<melvster>imagine we could have a global scraping repository that added semantic markup to pages like facebook, google etc.
19:02:53<PovAddict>rreck: huh? anchors use ID, and the name attribute is deprecated
19:03:00<rreck>i will just "paint with a broad brush" for now
19:03:25<rreck>PovAddict: shows how much i know
19:03:35<melvster>you dont need to wait for pages to go rdfa then, you just apply the transform
19:03:55<PovAddict>/page.php#info_section_contact would make the browser focus on that div
19:04:10<melvster>if someone wants a site to be semantic they just highlight the iteam a la autopager/firebug and type in suggested predicates
19:04:21<melvster>s/iteam/item
19:04:51<melvster>then the addon downloads the xml, and everyone else gets the benefit
19:04:56<rreck>so the assertions apply to only what is contained in the div tag?
19:05:34<rreck>inside that div tag i mean
19:05:58<rreck>that is a great answer
19:06:06<melvster>if using xpath i guess so, but i suspect someone here will know the 'proper' way of doing things
19:06:57<rreck>which is why i asked, seems like PovAddict has abundantly more clue than myself. thanks.
19:07:28<melvster>hmm i didnt know that either :)
19:07:59<PovAddict>people sometimes use <a name="part1"><h1>Part 1</h1>
19:08:09<PovAddict>which is even wrong because an <a> needs a closing tag
19:08:27<rreck>so DIV is much much better for that exact reason
19:08:38<PovAddict>then add some CSS rule that makes links show underlined when hovered, and problems happen (the title gets underlined when hovered even though it's not a link :P)
19:08:42<PovAddict>the right way would have been
19:08:55<PovAddict><h1 id="part1">Part 1</h1>
19:09:05<PovAddict>and you can still link to it the same way as before: page.htm#part1
19:09:33<rreck>but im trying to say something about what is contained inside those tags
19:09:56<PovAddict>I'm just saying how browsers work wrt that
19:10:06<rreck>perfect
19:10:09<PovAddict>in this channel I'm below average in RDF knowledge :)
19:10:24<rreck>i sure know THAT feeling
19:11:03<rreck>well it goes to reason that the anchor refers to a section, the assertions made wrt to the anchor have the same scope
19:11:18<PovAddict>so I don't know if making triples with page.htm#id_of_some_element_in_html_page as object is the way to give information of stuff inside that element
19:11:22<PovAddict>probably is
19:33:23<melvster>Microsoft's Bing overtook Yahoo! as the number two search engine in the United States and worldwide Thursday (4th June) according to data from StatCounter Global Stats.
19:34:00<PovAddict>what's Bing? :P
19:34:20<melvster>bing is not google ;)
19:34:29<PovAddict>!
19:37:20<melvster>Microsoft's Bing overtook Yahoo! as the number two search engine in the United States and worldwide Thursday (4th June) according to data from StatCounter Global Stats. StatCounter says that Bing grabbed market share from Google.
19:37:25<melvster>sorry
19:37:34<melvster>in the US Bing leapfrogged Yahoo to take second place on 16.28%. Yahoo! has 10.22%. Google still commands the US search engine market with 71.47%.
19:37:43<mattl>surely this is a blip?
19:37:53<mattl>someone should just buy yahoo anyway.
19:38:05<mattl>in a few months, i'll buy them for about $500
19:38:16<PovAddict>I'll Buy That For A Dollar
19:38:21<melvster>:) i picked them up when they were $9
19:38:42<mattl>no, the whole company for $500.
19:39:36<melvster>oh!
19:39:51<melvster>they have about 5 billion in cash, so you'd probably make a decent profit :)
19:41:16<melvster>i have to say google were by far and away the best search engine 10 years ago, but have good but not exceptional in the time since, regarding their search engine, maybe it's time the competition caught up
19:41:28<melvster>(been)
19:42:01<melvster>they've innovated in lots of other areas, though
19:42:14<PovAddict>company is "too big"...
19:43:49<melvster>its also hard when your mission is to organize the worlds data, there's a lot of data out there, i think facebook alone take 15 TB into their hadoop system each day
19:45:35<melvster>i think the first one of them to really understand distributed architecture, and semantic markup, and really do it in an open way, will lead the pack, perhaps yahoo is in front in that respect, its hard to say
19:51:25<melvster>but i do actually quite like bing
19:51:46<melvster>i did buy a product from their shopping tab withing a few minutes of using it
19:51:59<melvster>which ive never done with other search engines
20:45:29<csarven>ACTION is still unsure about what makes Bing so special
20:47:23<PovAddict>Bing Is Not Google

Back to channel and daily index: content-negotiated html turtle